alt Hacker NewsPost starts with wrong statement right away:
"The Transformer architecture revolutionized sequence modeling with its introduction of attention"
Attention was developed before transformers.
Post starts with wrong statement right away:
"The Transformer architecture revolutionized sequence modeling with its introduction of attention"
Attention was developed before transformers.
> Attention was developed before transformers.
I just looked this up and it’s true, this changes the timeline I had in my mind completely! I thought the paper on Transformers is what also introduced the attention mechanism, but it existed before too and was applied on RNN encoder-decoder. Wow